On October 14, the Goldhirsh Foundation partnered with Bloomberg to present “AI For Social Good,” the fourth LA Tech Week event organized by the Foundation during the past three years.

For the first time in months, Los Angeles woke up to thunder and lightning. Everyone knows the stereotype about Angelenos and driving in the rain, but on this Tuesday, folks didn’t just come out, they filled the room. Hosted at Bloomberg’s Century City office on the 31st floor, the foggy views were stunning – and the conversation was flowing. No storm could stop the social impact sector.

Bloomberg Managing Editor Mark Gurman – a highly regarded technology journalist – moderated the two-panel session, which began with a big-picture conversation on the AI landscape for the social impact sector featuring Tara Roth, president of the Goldhirsh Foundation, and Kendall Arthur, director of strategic partnerships at San Francisco-based Fast Forward.

Gurman then transitioned to a deeper dive about nonprofits putting AI into practice, featuring Goldhirsh Foundation AI Executive-in-Residence Jen García, Defy Ventures Southern California Executive Director Quan Huynh, and Develop for Good Founder Mary Zhu, who like Arthur, was in town from the Bay Area.

The Big Picture

Roth and Arthur explored the larger-scale ways nonprofits can utilize AI for good, and the gaps AI can fill to achieve further success. “We surveyed 600 impact organizations in Los Angeles to find out what their needs were,” Roth noted. While nearly 70 percent were interested in AI, “they didn’t know how to access it, and didn’t know what questions to ask.” Arthur received similar feedback in her work at Fast Forward, whose mission uses tech to solve major global issues.

The Goldhirsh Foundation is providing some of its portfolio companies, and Goldhirsh Foundation and Fast Forward are each providing smaller nonprofits and public good organizations, with the tools to leverage AI to maximize impact. Roth discussed how the Goldhirsh Foundation considers “AI as a way to help future-proof the organizations we work with,” and has provided more than 300 nonprofits with AI training and consulting through García and a collaboration with OpenAI. The Foundation additionally publishes AI For Impact blog posts, highlighting lessons learned and use cases.

Fast Forward, meanwhile, hosts a flagship accelerator program that supports nonprofits with original tech and ethical AI for humanity, assisting operations, fundraising, and more. “We lift them up, so they can achieve impact at scale,” Arthur said.

A Deeper Dive

García, Huynh, and Zhu took center stage next to discuss real-world AI use cases in nonprofits today. García broke down how she works with organizations to reach their full potential using AI, particularly in AI literacy training and skills development. She emphasized using AI to build one’s abilities; for instance, strengthening communication skills by asking the right questions: “‘How do I communicate this message differently?’ You learn from the tools, rather than cognitively offload onto them.”

Develop for Good pairs college students with nonprofits to simultaneously provide organizations with software and students with workplace experience. Zhu described how in a recent program, students used AI to create their own personal roadmaps to assist project planning; by the end, nearly half of students noted that they found them helpful in navigating their work. Develop for Good is now hiring their first full-time software engineer to develop this technology further, and make their program more scalable.

Defy Ventures is a multi-time LA2050 Grants Challenge winner focused on entrepreneurship for formerly incarcerated individuals. He has built custom GPTs to enhance and streamline daily communications and operations, and prep for fundraising pitches. (He credits his Donor Dialogue Coach GPT for helping earn a $30,000 grant.) Huynh also told the Tech Week crowd that he saw a real time-saving benefit from incorporating AI into his development work: “In writing grants,” he said, “I went from 20-30 hours to two hours.”

Empowerment, not Replacement

Several attendees said they were drawn to “AI For Social Good” because it addressed AI safeguards and ethics, rather than solely focusing on growth. AI safety and retaining human ability and creativity were major focuses of both panel sections.

Roth and Arthur discussed the cruciality of AI security, particularly as many of the nonprofits in the LA2050 and Fast Forward cohorts work directly with vulnerable populations or sensitive data. As Arthur put it, “How can we make sure that we’re protecting the people who are coming to these nonprofits for a need, without compromising the efficiency of the technology and the type of support they’re getting?”

García, Huynh, and Zhu all discussed the importance of using AI to enhance skills and ability in their organizations. Zhu urged intentionality, ensuring that AI is being optimized for the people organizations are committed to serving. “We want AI to be a tool that can increase students’ critical thinking and become effective problem solvers,” she said.

Huynh agreed, emphasizing that AI tools are leverage, not a hand-out. García stressed the need for AI values and writing for yourself – “writing is thinking,” she said, emphatically, and urged users to ask themselves, “Am I using this to learn? Is this beneficial to my audience?”

After the panels concluded, Roth delivered closing remarks, and attendees and participants stayed to connect, continue the conversation, and discuss potential collaborations. Energized and thoughtful, the full house eventually trickled out, emerging into what became a sunny day.

Photos, from top: 1) From left to right: García, Huynh, Gurman, Roth, Zhu, Arthur, and Bloomberg Corporate Philanthropy’s Sabrina Briefel; 2) Arthur and Gurman; 3) From left to right: García, Zhu, and Huynh. Photos courtesy Bloomberg.

[Click here for a collection of posts about the extensive work in AI by the Goldhirsh Foundation and its LA2050 initiative.]